Modeling Interactive Neural Processes

DESCRIPTION (provided by applicant): This research proposal seeks to bridge the gap between computational theories of executive motor control and the neural processes that underlie the control of movements. Often, qualitatively different computational theories can explain the same behavioral data. However, in some cases competing theories can be distinguished based on the form of the instantiating neuronal mechanisms. A primary research goal of this proposal is to elucidate the correspondence between a specific cognitive theory of control and a specific neural instantiation. To that end, behavioral and neural data obtained in a countermanding (stop signal) task will be linked through a simple neural network model constrained at once by the particular characteristics of countermanding behavior, the logic of the computational theory that accounts for the behavior, and the properties of neurons recorded while macaque monkeys perform the task. The training goal is to become conversant with the nuances of the neural circuits instantiating motor control and to develop expertise with stochastic models of human cognition and performance. The strength of this proposal lies in its unique convergence of conceptual frameworks and the breadth of data considered.